Financial stability risks from geoeconomic
fragmentation
January 2026
ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation
Financial stability risks from geoeconomic fragmentation
Contents 1
Contents
Executive summary 2
1 Introduction 6
2 A framework for assessing geopolitical risks to financial stability 7
3 Assessing geopolitical risks to EU countries 11
Classification of geopolitical risk indicators 11
Box 1 Geopolitical risk: from US-centric indices to euro area tools 16
Selection of informative indicators 18
Box 2 Recommended indicators for geopolitical risks analysis 19
Current state of geopolitical risks 22
4 Transmission of geopolitical risk 26
Macro-financial transmission of geopolitical risk 26
Tail risks to macro-financial conditions in the EU 35
EU country findings 49
5 Financial market impact of geopolitical risks 53
Geopolitical risk and financial market spillovers 53
EU financial market reaction to US political risk shocks 58
6 Impact of geopolitical risk on euro area financial institutions 62
Impact of the Russian invasion of Ukraine on euro area banks 62
US economic policy uncertainty and euro area bank lending 72
Geopolitical shocks and euro area sectoral portfolio reallocation 75
Box 3 The UCITS sector in Luxembourg in the face of the Russian
invasion of Ukraine 80
References 83
Annex 91
Financial stability risks from geoeconomic fragmentation
Executive summary
Executive summary
Geoeconomic fragmentation and geopolitical risk have emerged in recent
years as defining challenges for economies worldwide, profoundly reshaping
the global economic and financial landscape. Heightened geopolitical tensions,
trade frictions, and regulatory divergence have created a world where political and
economic spheres are increasingly intertwined. These developments have
transformed the nature of uncertainty faced by policymakers and financial
institutions, as geopolitical shocks transmit through global markets, disrupt trade and
investment flows, and potentially impact financial stability.
This report examines the link between geopolitical and geoeconomic risks and
financial stability. It introduces a comprehensive framework for regular monitoring
of geopolitical risks, integrating a wide array of quantitative indicators with a rich
empirical modelling foundation to evaluate the impact on the financial sector. The
framework is conceived as an instrument for policymakers and financial authorities,
enabling a systematic assessment of how geopolitical developments affect macro-
financial conditions. The work makes three main contributions to risk monitoring.
First, it establishes a monitoring toolkit for indicator-based analysis of geopolitical
risks that can be readily integrated into existing financial stability frameworks.
Second, it explores the macro-financial transmission of geopolitical shocks using
state-of-the-art econometric models. Third, it draws on granular datasets to
document how banks and non-banks adjust to geopolitical shocks and periods of
elevated policy uncertainty.
The framework categorizes geopolitical risks into five broad groups, each
capturing facets and channels that can amplify macro-financial vulnerabilities
and impact financial stability. The categories are: (i) military conflicts and wars; (ii)
infrastructure vulnerabilities; including energy and digital systems; (iii) trade
disruptions and sanctions; (iv) capital and financial risks; and (v) political or societal
factors. Together, these dimensions shape the complex landscape of geopolitical
risk. The framework also identifies the main channels through which these risks
affect financial stability – financial, real-economy, and operational. This report
focuses on the financial channel, where spillbacks occur through tighter financial
conditions, higher risk premia, and financial market stress. As a result, financial
institutions may face heightened credit, market, liquidity, or operational risks.
Moreover, existing vulnerabilities and feedback loops can act as amplifiers,
transforming localised disturbances into systemic risks with the potential to threaten
financial stability.
The empirical analysis relies on a broad set of 40 geopolitical indicators
covering the five risk categories identified in the framework. These metrics
include high-frequency market-based measures and slower-moving cyclical and
structural indicators. Statistical and econometric selection criteria were used to
identify the most relevant indicators for visual monitoring tools and econometric
analysis. The geopolitical indicators heatmap (GEO heatmap) can be embedded in
Financial stability risks from geoeconomic fragmentation
Executive summary
broader risk monitoring frameworks to support ongoing financial stability
assessments.
The monitoring tools used in the analysis confirm that the prevailing
geopolitical risks have intensified in recent years, as captured by trends in
geoeconomic fragmentation, geopolitical tensions and heightened policy
uncertainty. Measures of policy uncertainty have especially surged during 2024 and
2025, driven primarily by a sharp increase in global economic and trade policy
uncertainty. These developments have been accompanied by a 27% rise in trade
disputes at the World Trade Organization between 2015 and 2024 combined with
regulatory divergence.
A significant dichotomy emerges between a measured rise of geopolitical risk
and the impact for the economy and financial stability. While indicators of
uncertainty have surged and model-based results indicate substantial downside risks
for the real economy, measures of financial volatility have remained contained or
quickly reverted after short-lived spikes. Growth-at-Risk (GaR) estimates – capturing
real GDP growth at the lower tenth percentile – show that the inclusion of geopolitical
indicators lowers expected growth outcomes compared with estimations without
such risk factors. Since 2014, the contribution of geopolitical indicators to these
dynamics has increasingly acted as a drag on GaR, reducing it by one to two
percentage points. On the financial side, heightened geopolitical shocks and policy
uncertainty tends to raise systemic stress, resulting in lower loan growth and tighter
lending conditions. These results point to significant tail risks for the real economy
from geopolitical sources.
The transmission of geopolitical risks and economic policy uncertainty varies
notably across EU Member States, reflecting differing levels of risk exposures
and financial sector resilience. Based on the empirical evidence, economies with
higher trade openness or higher public indebtedness appear more vulnerable to
amplification effects. The evidence also shows that geopolitical risk can reduce
financial integration and economic synchronisation within the euro area, amplifying
output losses and financial stress. While heterogeneous exposure to geopolitical risk
among EU Member States can partly explain these divergences, it suggests that
reduced synchronisation and financial integration may compound the macro-financial
effects of geopolitical shocks beyond their direct impact.
To evaluate the resilience of financial systems, the empirical analysis is
complemented with sensitivity analyses and macro-financial stress tests.
These stylized scenario exercises illustrate how macro-financial variables and risks
surrounding them evolve under simulated geopolitical shocks. In addition, they
identify the contribution of the financial system as an amplifier of the dynamics and
the benefits of stabilising financial stress to mitigate the effects. Taken together, the
results highlight the stabilising role that a sound and well-capitalised financial system
can play in limiting adverse macro-financial outcomes.
Financial markets provide a critical lens through which to observe how
geopolitical shocks propagate and affect interlinkages between market
segments. Volatility spillovers between asset classes – bonds, commodities,
Financial stability risks from geoeconomic fragmentation
Executive summary
equities, and exchange rates – spiked during major recent events such as the
COVID-19 pandemic and the Russian invasion of Ukraine in 2022. These episodes
illustrate how geopolitical shocks can break down established relationships between
markets and significantly impact cross-market interconnectedness. An additional
analysis focuses on the transmission of US political risk. It is based on prediction
market data, which are designed to track the emergence of new geopolitical risks
and to analyse their impacts across financial markets. The analysis shows that
political risk shocks linked to US trade tariffs not only affect US equity prices and
exchange rates but also spill over into euro area equity markets.
A granular perspective on financial institutions confirms that both banks and
non-banks adjust their behaviour in response to geopolitical shocks by
reducing lending, especially across borders. The effect on lending was strongest
among euro area banks with less capital headroom or large exposures to high-risk
countries. Banks reduced both the probability of new lending relationships (by
around 6%) and the average loan amount (by 9%). Similarly, spillovers from
heightened US policy uncertainty led to a reduction in credit from euro area banks.
On the liability side of banks’ balance sheets, market-based funding tends to decline
in response to increases in economic and trade policy uncertainty and rising
geopolitical risk, particularly as regards foreign-currency funding. For non-bank
financial institutions, the shift towards domestic concentration following the Russian
invasion of Ukraine was pronounced, with exposures to the rest of the world falling
by 17%, compared with 7% for domestic exposures, driven primarily by investment
funds, pension funds and insurers. This re-orientation towards domestic markets –
as opposed to a rebalancing from higher to lower-risk cross-border exposures –
reduces exposure to external shocks but may also constrain diversification and the
system’s ability to mitigate risk.
While this report provides a comprehensive assessment of the financial
stability implications of geopolitical risk and geoeconomic fragmentation,
certain gaps remain. The availability and comparability of geopolitical indicators
vary considerably over time and across countries, suggesting the need for further
efforts to build methodologically harmonised datasets and tailor monitoring
frameworks to national specificities by selecting specific indicators for different
countries. Moreover, new sources of risk are emerging rapidly, including digital
fragmentation, cyber threats, and the societal dimensions of geopolitical shifts.
Economic, financial, and regulatory divergence – whether through trade restrictions
on critical materials, financial regulation of new instruments, or climate-related policy
shifts – may further complicate the landscape. It could therefore be worthwhile to
complement indicator-based monitoring with more elaborate scenario analyses
involving specific risk materialisations, as they would enhance the capacity to
anticipate emerging geopolitical and financial stability risks.
By offering a structured, empirically grounded approach to understanding the
financial stability implications of geopolitical risks, the monitoring toolkit
presented in this report provides a foundation for future research and policy
development. Policymakers and financial institutions can employ this framework to
detect and categorize the main sources of risk in an evolving landscape of
Financial stability risks from geoeconomic fragmentation
Executive summary
geopolitical risk to assess cross-market and cross-border spillovers, estimate the
impact on financial institutions and calibrate macro-prudential responses. In an era of
accelerating fragmentation and persistent geopolitical uncertainty, proactive risk
management and strengthened cooperation across jurisdictions are essential to
preserve financial stability and sustain economic resilience.
Financial stability risks from geoeconomic fragmentation
Introduction
1 Introduction
Geoeconomic fragmentation and geopolitical risks have emerged as critical
challenges for economies worldwide, reshaping the global economic and
financial landscape in profound and interconnected ways. The era of
globalisation in the late 20th and early 21st centuries created strong
interdependencies between the EU and global economies, including its financial
system. Ongoing shifts in policy priorities and regulatory frameworks are not only
affecting trade and global supply chains, but also giving rise to risks and
uncertainties that reverberate across societal and political structures, impacting
financial systems and their stability.
This report provides a comprehensive toolkit for analysing the relationship
between financial stability, geopolitical risks and uncertainty. It offers a
framework for understanding the transmission of different types of geopolitical and
geoeconomic risks with direct and indirect effects on the EU’s economy and its
financial system. To effectively measure and monitor the risks associated with
geopolitical developments, including geoeconomic fragmentation, a comprehensive
database has been created, encompassing a wide range of indicators organised by
geopolitical dimension. The indicators were selected to achieve a broad view of
geopolitical risk, combining high-frequency market signals with slower-moving
structural trends. This approach ensures thematic coverage and empirical relevance
for assessing the transmission of geopolitical developments and provides a
monitoring framework for the Eurosystem and the European Systemic Risk Board
(ESRB).
Building on a robust empirical analysis, the report identifies and assesses
several transmission channels through which geopolitical risks propagate to
financial systems. The suite of models used includes a range of geopolitical
indicators and analyses their transmission to the financial system, along with their
implications for financial stability. It uses microeconometric and time series analysis,
such as structural vector autoregressions (VARs) or factor-augmented VARs, to
assess the impact of geopolitical risks on the main macro-financial developments.
The report is structured as follows. Section 2 presents the conceptual framework,
providing a detailed description of the drivers and transmission channels and offering
insights into the mechanisms that link geopolitical events to financial stability.
Section 3 covers a rich set of indicators along the main geopolitical risk categories,
namely military conflicts, infrastructure, trade, capital and finance, as well as political
and societal factors. Section 4 provides empirical analyses quantifying the impact of
geopolitical shocks on the euro area countries and on other EU Member States,
including the risks facing their economies and financial systems. Section 5 focuses
on transmission and spillovers within the financial system, while Section 6 examines
the responses of financial institutions, including both the banking sector and non-
banking financial institutions, to geopolitical risks.
Financial stability risks from geoeconomic fragmentation
A framework for assessing geopolitical risks to financial stability
2 A framework for assessing geopolitical
risks to financial stability
This report sees geopolitics as a broad concept, encompassing the use of all
available instruments by nation states to pursue their interests and expand
their power and influence. These instruments include military strength, as Caldara
and Iacoviello (2022) define geopolitical risk as “the threat, realization, and
escalation of adverse events associated with wars, terrorism, and any tensions
among states and political actors that affect the peaceful course of international
relations”. Moreover, geopolitics also involves instruments of hybrid warfare, such as
cyberattacks, physical sabotage and the spread of disinformation. Additionally,
geopolitical objectives are often pursued through economic means, including trade
policies, sanctions and currency manipulation – collectively referred to as
geoeconomic instruments. By adopting a broad definition of geopolitics, this report
also considers geoeconomic
The primary aim of this report is to analyse the impact of geopolitical risk on
financial stability. This includes examining the effects of sudden geoeconomic
shocks, as well as the gradual process of geoeconomic fragmentation, which Aiyar
et al. (2023) define as “a policy-driven reversal of global economic integration often
guided by strategic considerations”. Geoeconomic fragmentation can therefore be
seen as a form of strategic disintegration driven by geopolitical motives, as defined
by Mohr and Trebesch (2025).
The framework encompasses the key transmission channels through which
geopolitical and, more specifically, geoeconomic risks affect financial stability.
This includes the amplifiers and dampeners, and the effects on capital flows,
financial markets and institutions. This section first describes the shock and trend
characteristics of geopolitical risks and divides them into five different categories.
Second, it sets out the impact of geopolitical risks on uncertainty and volatility, as
well as the way in which this impact is propagated through various transmission
channels. Third, it discusses the risks to financial institutions, the amplification and
damping mechanisms, as well as the impact on financial stability.
Geopolitical risks may arise from the materialisation of shocks that entail a
slow-moving structural rise in fragmentation. Sudden materialisation may take
the form of unexpected shocks, such as the start of a military conflict, whereas
fragmentation develops as a trend or pattern, potentially multifaceted, over
prolonged periods of non-cooperation, sanctioning or coercion. Recent events linked
to fragmentation include Brexit and the COVID-19 pandemic.
The analysis in this report combines several dimensions of geopolitical risks,
encompassing a wide range of events and actions that can undermine
1 A more detailed description of the concepts of geopolitics and geoeconomics can be found in the
Annex 1 to this report.
Financial stability risks from geoeconomic fragmentation
A framework for assessing geopolitical risks to financial stability
political, economic and financial stability. The different types of geopolitical risks
fall into five different categories: (i) military conflict and war; (ii) infrastructure; (iii)
trade; (iv) capital and finance; and (v) politics and society (Figure 1).2 These risks
can have a profound impact on financial institutions, markets and, more broadly,
financial stability.
Figure 1
Categorisation of geopolitical risks
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Categorising geopolitical risks ensures clearer identification of transmission
mechanisms and potential spillovers. Indirect effects on financial stability often
emerge through trade or political channels. Escalating geopolitical tensions may, for
instance, lead to protectionist policies that fragment global trade, reduce economic
growth and increase uncertainty, affecting credit conditions and asset valuations.
Direct effects, by contrast, may stem from sudden shifts in capital flows or
heightened risk perception during military conflicts. An unexpected invasion may, for
example, trigger a flight-to-safety, causing sharp movements in exchange rates and
bond yields, and tightening global financial conditions.
The transmission channels of shocks may induce uncertainty, on the one
hand, and volatility, on the other. Uncertainty arises from a lack of predictability as
regards future geopolitical developments and their outcomes, affecting expectations
and increasing risk aversion among businesses, households and market
participants. Volatility refers to verifiable rapid and unpredictable changes in, for
example, asset prices, trade and capital flows, commodity prices, foreign exchange
rates, energy prices, interest rates and risk spreads. It is useful to conceptually
differentiate these two aspects given that this helps clarify how geopolitical risks
influence the behaviour of economic agents and financial markets. Uncertainty is a
more general phenomenon related to the behaviour of economic agents in the face
of unknown events, whereas volatility is an observable measure for assessing
deviations from historical averages of financial or economic variables.
Geopolitical risks in general, and geopolitical shocks in particular, can have
particularly strong effects and produce an unusual degree of volatility and
uncertainty among economic agents, and for the real economy and financial
2 A more detailed description of the different types of geopolitical risks can be found in the Annex 1.
Financial stability risks from geoeconomic fragmentation
A framework for assessing geopolitical risks to financial stability
markets. Given the nature of geopolitical shocks, they seem to produce an
unusually high degree of volatility and uncertainty which is often far removed from
economic market knowledge and information. Unlike standard economic
disturbances, geopolitical shocks often originate outside the economic system and
are amplified through the weaponisation of interdependencies – for example,
restricting access to energy, raw materials or semiconductors to exert geopolitical
pressure. They tend to trigger sudden supply disruptions, as illustrated by Russia’s
drastic reduction of gas supplies to the EU in 2021-22, which led to a surge in
inflation. Moreover, geopolitical tensions can undermine confidence in financial
markets and erode trust in global trade rules, creating systemic uncertainty.
A fragmentation of trade along geopolitical lines would exert significant and
persistent effects on real GDP, trade, and inflation (International Relations
Committee, 2024). Trade fragmentation is perhaps the best-known example of the
materialisation of geopolitical shocks. Model-based simulations show that in a world
economy fragmented along geopolitically opposed blocs, real output would be
durably lower. This reflects the loss of efficiency arising from the breaking up of
global value chains, with spillovers particularly strong in highly interconnected
economies. Global trade volumes would decline sharply, given that barriers and
frictions not only reduce bilateral flows but also distort the structure of production. A
shock to trade fragmentation would cause persistent inflationary pressures for
countries imposing higher barriers to trade, given that the economy only
progressively adjusts to reduced competition, higher production costs and strategic
restrictions on key inputs. In addition, counterfactual analysis shows that a
fragmented world economy would face larger and more frequent supply shocks,
triggering higher output and inflation volatility given that a fragmentation of supply
chains reduces the ability of economies to absorb shocks. Finally, even if no trade
barriers are put in place, the mere perception of trade policy uncertainty can weigh
on macroeconomic outcomes (Caldara et al., 2020). This is because uncertainty
induces precautionary behaviour, such as firms delaying investment and households
increasing precautionary savings, which depresses growth and trade dynamics. This
uncertainty also feeds into financial markets, raising risk premiums and tightening
financial conditions, thereby amplifying its real economic effects.
Beyond the real and operational channels, the transmission of geopolitical
risks can be particularly strong through the financial channel, owing to the
uncertainty and volatility these risks While the real and operational
channels generally give rise to indirect risks for the financial system, the financial
channel exerts a more immediate and direct influence on the system, making this
channel particularly important for financial stability assessments (Figure 2).
Transmission through the financial channel can be analysed effectively through the
dual lens of uncertainty and volatility. Uncertainty undermines investor sentiment and
heightens risk aversion, leading to higher risk premiums and precautionary liquidity
hoarding. Volatility amplifies these effects by triggering market sell-offs and
3 This concept is in line with the transmission channels identified by ECB Banking Supervision – the
financial market channel, the real economy channel, and the safety and security channel – in assessing
the impact of geopolitical risk on credit, liquidity, market, business model, operational risks and
governance. See Buch (2024) and the article entitled “Addressing the impact of geopolitical risk” on the
ECB website.
Financial stability risks from geoeconomic fragmentation
A framework for assessing geopolitical risks to financial stability
heightening stress in key funding markets. Together, these dynamics drive flight-to-
safety behaviour, sharp capital flows, fluctuations in foreign exchange rates and risk
repricing in asset markets. The resulting wider spreads and credit tightening – often
intensified by fire sales – result in a deterioration in financial conditions and elevated
default rates, ultimately straining the balance sheets of financial institutions.
Figure 2
EU geopolitical risks and fragmentation analysis framework
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Note: The transmission channels in the dashed box were not the focus of the analysis conducted by the ECB/ESRB workstream on
financial stability risks from geoeconomic fragmentation.
The prominence of the financial channel for macroprudential analyses has
important implications for risk measurement and indicators. Financial stability
analyses focus on the (valuation) effects on the balance sheets, solvency and
liquidity of financial institutions. In addition, consideration of bilateral investment
positions is appropriate in order to accurately assess spillover risks and cross-border
effects. Feedback loops and interactions between the financial, real and operational
channels may amplify the initial impact of geopolitical shocks, especially if they
interact with vulnerabilities in the financial system.
In addition to broader market dynamics, geopolitical risks generate specific
vulnerabilities for financial institutions. Rising default rates and falling asset
prices increase credit and market risk, while funding stress and margin calls heighten
liquidity risk. Cyberattacks and operational disruptions further amplify operational
risk. The transmission of these risks may be intensified by financial contagion,
inadequate policy responses and structural fragilities, such as high leverage or
concentrated exposures.
The risks to financial institutions in combination with amplifying factors could
culminate in systemic risks that pose a threat to financial stability. Financial
distress can spread across institutions and markets, leading to severe disruptions
and a loss of confidence in the financial system. Financial institutions may struggle to
perform their intermediary functions effectively and face threats to their business
model, with an adverse impact on the provision of credit and financial services to the
real economy. Persistent financial stress can further delay recovery and erode trust
in the financial system, leading to bank runs, capital flight and further instability that
may ultimately require the intervention of public authorities.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
3 Assessing geopolitical risks to EU
countries
Classification of geopolitical risk indicators
Monitoring and analysing geopolitical risks requires a coherent set of
indicators that are both conceptually meaningful and empirically relevant. The
selection presented in this section is designed to obtain a broad view of geopolitical
risks and their interaction with macro-financial and structural vulnerabilities in EU
Member States. The selection of indicators ensures broad thematic coverage and
combines well-established measures with more novel proxies, including high-
frequency market-based signals and slower-moving structural variables, to track
sudden developments as well as persistent trends.
The indicator set covers all the categories in the framework set out in Section
2 and adds a general risks category. The wide-ranging set of indicators captures
geopolitical risks in each of the framework’s topical categories and also includes a
category covering general risks that groups together broad, global measures
capturing uncertainty and volatility. The full classification therefore consists of six
categories: (i) general; (ii) military conflicts and wars; (iii) infrastructure; (iv) trade; (v)
capital and finance; and (vi) politics and society.
In addition to thematic grouping, the indicators are also classified by the
nature of the signal they provide. Building on the definitions given in Section 2,
each indicator was assigned to uncertainty, volatility or trend. Uncertainty measures
capture ex ante risk perceptions, typically reflecting changes in sentiment before
shocks materialise, but may themselves have economic and financial effects
(Sections 4 and 5). Volatility indicators represent the realised impact of events, often
visible in financial market data or other high-frequency datasets. Trend indicators
capture slow-moving or structural shifts, such as long-term changes in trade
openness, fragmentation or defence spending. Figure 3 provides a visual
representation of this classification that links the six thematic categories (shown on
the left of the chart) to the individual indicators (shown in the centre of the chart) and
to the three signal types (shown on the right of the chart). This mapping shows how
certain indicators fit into multiple categories or combine multiple signal properties.
The indicators selected cover a diverse range of sources and methodologies,
which is critical for interpreting their behaviour. The set of indicators covers long
time periods – for some indicators, starting in the 1960s. It comprises 17 news-based
indices, 12 model-based measures, three composite indicators, six indicators from
official statistics and four events-based measures (Table 1). News-based indices,
such as the Geopolitical Risk (GPR) Index, the Economic Policy Uncertainty (EPU)
Index, the World Trade Uncertainty Index (WTUI) and the Sanctions Intensity Index
(GSI), rely on media coverage and therefore measure geopolitical events indirectly,
based on the focus of the selected media source. Model-based measures, including
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
the Jurado, Ludvigson and Ng (JLN) Macroeconomic Uncertainty indices – Real
Uncertainty (RealUn), Macro Uncertainty (MacroUn), Financial Uncertainty (FinUn) –
and the Common Volatility (COVOL) are derived from model-based estimation and
capture underlying risk dynamics across different frequencies. Composite indicators,
such as the Global Supply Chain Pressure Index (GSCPI), combine multiple inputs
into a single measure, providing a broader perspective on risk conditions. Indicators
from official statistics, such as the Military Expenditure, Trade Openness (TO) and
the Allocated Foreign Exchange Reserves (FXR) Indicators, offer slow-moving
structural information anchored in official, internationally comparable data (see also
Box C). The events-based category includes the Number of Conflict Events
(NoConfl) Indicator, the Number of International Armed Conflicts (NoArmedConfl)
Indicator and the Global Number of Active Sanctions (NoSanct) Indicator, which
directly record the occurrence of specific geopolitical events or measures.
Recognising these distinctions is essential for interpretation and relevance when
integrating such indicators into monitoring tools and empirical models. Beyond the
individual indicators, it may be necessary to explicitly account for the exposure to the
risks, relevant in . the transmission channel through cross-border financial
positions, discussed in more detail in Box in the Annex.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Figure 3
Categorisation of geopolitical risk indicators for the EU geopolitical risks analysis
framework
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Notes: GPR stands for geopolitical risk and EMV for equity market volatility. The categories on the left follow those of the framework
used in this report (Section 2) To reflect the different types of uncertainty and volatility, the indicators are grouped into the following
categories: uncertainty (ex ante), volatility (materialised) and trend development. Indicators between stars (* … *) are those included in
the geopolitical indicators heatmap, while the remainder serve for more targeted analytical purposes, such as trend detection or
robustness checks.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Table 1
Indicators of geopolitical risk and fragmentation for the EU geopolitical risks analysis
framework
Indicator Short Uncertainty/
Volatility
Trend Source/
Authors
Frequency Type Use in
empirical
analysis
General
1 World Uncertainty Index WUI ✓
Ahir, Bloom & Furceri
(2018)
Quarterly News-based ✓
2 Economic Policy Uncertainty
Index
EPU ✓
Baker, Bloom & Davis
(2016)
Monthly News-based ✓
3 JLN Real Uncertainty Index RealUn ✓
Ludvigson, Ma, Ng
(2021)
Monthly Model-based
4 JLN Macro Uncertainty
Index
MacroUn ✓
Jurado, Ludvigson & Ng
(2015)
Monthly Model-based
5 Geopolitical Fragmentation
Index & country bloc
subindexes
Frag
✓ Fernández-Villaverde et
al (2024)
Quarterly Model-based ✓
Military
6 Geopolitical Risk Index GPR ✓
Caldara & Iacoviello
(2022)
Monthly News-based ✓
7 GPR Threats Index GPRT ✓
Caldara & Iacoviello
(2022)
Monthly News-based ✓
8 GPR Acts Index GPRA ✓
Caldara & Iacoviello
(2022)
Monthly News-based ✓
9 Bondarenko Geopolitical
Risk Perceptions Indicator
GPRP ✓
Bondarenko et al.
(2024)
Monthly News-based ✓
10 Bilateral Indicator of Local
Perception of Geopolitical
Risk Indicator & by source
country or country groups
BiGPR ✓
Alonso-Alvarez et al.
(2025)
Monthly News-based
11 National Security Policy
EMV Tracker
EMVsec ✓
Baker et al. (2019) Monthly News-based
12 Number of Conflict Events
Indicator
NoConfl
✓ UCDP/GED Monthly Events-
based
13 Number of International
Armed Conflicts Indicator
NoArme
dConfl
✓ UCDP/PRIO Armed
Conflict Dataset
Monthly Events-
based
14 Military Expenditure
Indicator
MilExp
✓ World Bank Group Annual Official
statistics
Infrastructure
15 Significant Cyber Incidents
Indicator
NoCyb
✓ CSIS Quarterly Official
statistics
16 Energy-related Uncertainty
Index
EUI ✓
Dang et al. (2023) Monthly News-based
Trade
17 Trade Openness Indicator TO ✓ ✓ ECB data portal, MNA
Database
Quarterly Official
statistics
✓
18 Global Supply Chain
Pressure Index
GSCPI ✓
Federal Reserve Bank
of New York
Monthly Composite ✓
19 Trade Policy Uncertainty
Index
TPU ✓
Caldara et al. (2019) Monthly News-based ✓
20 World Trade Uncertainty
Index
WTUI ✓
Ahir, Bloom & Furceri
(2019)
Quarterly News-based
21 Trade Policy EMV Tracker EMVtp ✓
Baker et al. (2019) Monthly News-based ✓
22 Trade Fragmentation Index TFrag
✓ Fernández-Villaverde et
al (2024)
Quarterly Model-based
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Indicator Short Uncertainty/
Volatility
Trend Source/
Authors
Frequency Type Use in
empirical
analysis
Capital & Finance
23 Capital Restriction Index CC
✓ Fernandez et al. (2015) Annual Official
statistics
24 JLN Financial Uncertainty
Index
FinUn ✓
Jurado, Ludvigson & Ng
(2015)
Monthly Model-based ✓
25 Common Volatility Index COVOL ✓
Engle & Campos-
Martins (2023)
Daily Model-based ✓
26 Financial Fragmentation
Index
FinFrag
✓ Fernández-Villaverde et
al (2024)
Quarterly Model-based
27 Foreign Direct Investment
Ratio
FDI
✓ IMF BoP database,
ECB data portal
Quarterly Official
statistics
28 Financial Flows Ratio FinFlow
✓ IMF BoP database,
ECB data portal
Quarterly Official
statistics
29 Equity Market Segmentation
Index
EqSegm
✓ Bekaert et al. (2011) Quarterly Official
statistics
30 Allocated Foreign Exchange
Reserves Indicator
FXR
✓ IMF COFER Database Quarterly Official
statistics
Politics & Societal
31 Global Number of Active
Sanctions Indicator
NoSanct
✓ Global Sanctions
Database (GSDB),
Felbermayr et al. (2020)
Annual Events-
based
32 Sanctions Intensity Index GSI ✓
Bondarenko et al.
(2024)
Monthly News-based
33 Migration Fear Index MigFear ✓
Baker, Bloom & Davis
(2016)
Quarterly News-based ✓
34 Migration Policy Uncertainty
Index
MigPU ✓
Baker, Bloom & Davis
(2016)
Quarterly News-based ✓
35 Elections & Political
Governance EMV Tracker
EMVelec ✓
Baker et al. (2019) Monthly News-based
36 UN Votes Ideal Points
Indicator
UNideal
P
✓ Bailey, Strezhnev &
Voeten (2017)
Annual Model-based
37 Mobility Fragmentation
Index
MobFrag
✓ Fernández-Villaverde et
al (2024)
Quarterly Model-based ✓
38 Political Fragmentation
Index
PolFrag
✓ Fernández-Villaverde et
al (2024)
Quarterly Model-based ✓
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Note: EMV stands for equity market volatility.
In addition to the geopolitical risk indicators described above, the dataset
includes country-level and EU or euro area-level indicators that capture
vulnerabilities, systemic financial stress and the degree of market integration.
These indicators are not themselves measures of geopolitical risk, but rather
financial sector measures of transmission or outcome and can be viewed as
response indicators for the purposes of econometric analysis (Table 2). They allow
to assess how geopolitical shocks propagate through domestic financial systems,
interact with macro-financial conditions and influence the degree of EU financial
integration. The indicators include composite indices of domestic financial conditions,
such as the ECB’s Systemic Risk Indicator (SRI) and Financial Cycle (FinC)
Indicator. Volatility proxies, such as the ECB’s Country-Level Financial Stress
(CLIFS) Index and Composite Indicator of Systemic Stress (CISS), complement the
structural measures by capturing short-term changes in financial markets. Finally,
the ECB’s price-based and quantity-based financial integration indices provide a
structural perspective on capital market integration within the euro area.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Table 2
Overview of EU domestic trend and volatility indicators
Indicator Abbreviation Uncertainty/
Volatility
Trend/
Risk
level
Source/
Authors
Frequency Type Use in
empirical
analysis
i Common Composite
Indicator
CCI
✓ Constructed by
authors; SRI-
based
Quarterly Composite ✓
ii Systemic Risk Indicator SRI
✓ Lang, Izzo, Fahr,
Ruzicka (2019)
Quarterly Composite ✓
iii Financial Cycle Indicator FinC
✓ Schüler, Hiebert,
Peltonen (2020)
Quarterly Composite
iv Country-Level Financial
Stress Index
CLIFS ✓
Hollo et al. (2012) Monthly Composite ✓
v Composite Indicator of
Systemic Stress
CISS ✓
Hollo, Kremer &
Lo Duca (2012)
Weekly Composite ✓
vi CBOE Volatility Index VIX ✓
CBOE Daily Model-
based
vii Euro Stoxx 50 Volatility
Index
VSTOXX ✓
STOXX Daily Model-
based
✓
viii Equity Market Volatility
Index
EMV ✓
Baker, Bloom &
Davis (2019)
Monthly News-
based
✓
ix Price-based Financial
Integration Indicator
PriceFinInt
✓ Hoffmann, Kremer,
Zaharia (2019)
Quarterly Composite ✓
x Quantity-based Financial
Integration Indicator
QuantFinInt
✓ Hoffmann, Kremer,
Zaharia (2019)
Quarterly Composite ✓
Box 1
Geopolitical risk: from US-centric indices to euro
area tools
Geopolitical risks have grown more prominent in recent years, as highlighted
by the war in Ukraine and the ongoing conflict in the Middle East. Measuring
such risks is crucial for understanding their macroeconomic and financial
impact. Several approaches to the quantification of geopolitical risk have emerged
that use news-based indicators. This box compares three prominent indicators: the
Geopolitical Risk (GPR) Index (Caldara and Iacoviello, 2022), the Euro Area-specific
Geopolitical Risk (EA GPR) Index (Bondarenko et al., 2025) and the local perception
of Geopolitical Risk, based on the bilateral Geopolitical Risk (BiGPR) Index approach
focusing on local perception of geopolitical risk (Alonso-Alvarez et al., 2025).
The GPR Index reflects a US-centric Anglosphere perspective, limiting its
applicability to euro area analyses. The index uses keyword searches for terms
such as war, terrorism or geopolitical tensions in newspapers from the United States,
Canada and the United Kingdom. The authors developed both the GPR Index and
country-specific geopolitical risk indices. The latter, however, still reflect risk as
perceived through the lens of Anglosphere media. As a result, these indices are less
suitable for euro area-focused analyses.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
To address the limitation arising from the US-centric indicators, a monthly EA
GPR Index was constructed in Bondarenko et al. (2025). It is based on local
newspapers from five EU countries: Germany, France, Italy, Spain and the
Netherlands. The Index uses country-specific keyword sets in local languages and
was adapted from the original GPR phrase and is GDP-weighted.
The EA GPR Index captures regional risk perceptions more effectively than the
original (GPR) Anglosphere index. A comparison the two indices reveals similar
spikes during major global events (. following the 9/11 terrorist attacks and during
the Iraq War), but the EA GPR Index reacts more strongly to regionally proximate
shocks, such as the Russian invasion of Ukraine and the 2023 Gaza conflict (Chart
A). Importantly, the EA GPR Index has remained high since 2022, while the
Anglosphere GPR Index has returned to lower levels. This suggests regional
divergence in risk perception, likely to be driven by the proximity of the Russian-
Ukraine conflict. Moreover, empirical evidence suggests that euro area geopolitical
risk shocks affect euro area inflation, whereas Anglosphere geopolitical risk shocks
have less impact (although this is not shown here).
Chart A
Geopolitical risk indices
(Indicator units)
Source: Bondarenko Y., V. Lewis, M. Rottner and Y. Schüler (2025).
Notes: GPR stands for geopolitical risk. The Euro Area GPR Index is based on the index developed in Bondarenko et al. (2024). GPR
Anglosphere relates to the standard index developed in Caldara and Iacoviello (2022).
Another novel extension is the Bilateral Geopolitical Risk (BiGPR) Index
proposed in Alonso-Alvarez et al. (2025). It considers the local, country-level
perspective, but allows for directionality, with indices that measure how a specific
country perceives the risk associated with another country or region. This
directionality is crucial, given that the structural vector autoregression (SVAR) model
applied by the authors confirms that geopolitical shocks originating from different
regions tend to have different effects on different domestic economies.
The bilateral framework can also be used to break down the geopolitical risk
perceived by an individual country into its components. The national geopolitical
risk profile for the EU set out below (Chart B) shows that developments involving the
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
EU’s geopolitical partners (the western bloc) consistently account for the largest
share of EU geopolitical risk. The Middle East constitutes the second most important
component; this was especially true in 2011 with the Arab Spring. The breakdown
also highlights the importance of Russia’s actions for the wider geopolitical arena:
while the direct Russia component peaks during major events, such as the 2014
Crimea crisis and the 2022 invasion of Ukraine, its broader implications are reflected
in the substantial directional Russia-western bloc share of euro area geopolitical risk.
Chart B
National geopolitical risk profile for the EU
Note: The chart shows the top non-overlapping components of the external part of local geopolitical risk (computed for Germany,
Spain, France and Italy and weighted by GDP for 2023, with a one-year rolling average).
This decomposition reflects how the press attention to external geopolitical risk is
allocation across the different regions. Russia´s share surged in 2022 when about a
third of all external GPR references focused on the region, but has since then fallen
to roughly 8%. This decline primarily reflects media fatigue, a decrease in the
marginal value of news once the conflict becomes a persistent drop, even though the
underlying tensions may remain high. On the other hand, the consistently large
western-bloc component captures reporting on how the allied countries interpret and
coordinate their responses to geopolitical events, rather than the perception that
these allies are themselves a source of risk.
Selection of informative indicators
A structured process guided the selection of the most informative indicators.
The initial set of indicators covers a broad spectrum of geopolitical risks and
uncertainty, but many series are highly correlated or measure similar dynamics using
alternative data sources or methods. To reduce redundancy and focus on those with
the strongest analytical value, three main selection criteria are applied: (i) statistical:
frequent selection or strong loading across multiple statistical methods; (ii) topical:
clear alignment with relevant geopolitical risk categories and transmission channels;
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
and (iii) operational: practical attributes, such as timeliness, length of time series,
consistent cross-country coverage and ease of interpretation.
Multiple statistical and econometric methods supported the identification of
core indicators. The process has similarities to the analysis in Hodula et al. (2024)
and combines correlation analysis, Granger causality tests with macro-financial
variables, principal component analysis (PCA), hierarchical clustering and least
absolute shrinkage and selection operator (LASSO)-based variable selection in a
GaR Each method provides a different perspective for identifying
related indicators and highlights those with greatest empirical relevance for
assessing geopolitical risk. The full diagnostics, robustness checks and
methodological details are provided in The Annex.
Building on the empirical screening and diagnostic work, it is possible to
identify the indicators to be retained for more detailed analysis. The indicators
selected are described in Box B and are considered to be the most informative for
capturing the different dimensions of geopolitical risk and fragmentation. They are
used to populate the geopolitical indicators heatmap referred to in Section and as
inputs into the GaR time series and quantile vector autoregression (QVAR)
estimations (Section 4).
Box 2
Recommended indicators for geopolitical risks
analysis
The indicator shortlist balances high-frequency and structural measures
across signal Most of the indicators selected capture uncertainty or
volatility at daily, monthly or quarterly frequency, while a smaller set reflects slow-
moving structural trends at lower frequency. The balance achieved also ensures
representation across categories. The diversity of the indicators selected supports a
comprehensive assessment of both immediate shocks and persistent vulnerabilities.
General geopolitical uncertainty indicators – The Economic Policy Uncertainty
(EPU) Index6 is based on the frequency of policy-related uncertainty terms in
newspapers. In GaR applications, the global EPU Index shows a predictive value at
short horizons of up to one year, particularly for large EU economies. The JLN Real
Uncertainty (RealUn) Index is a model-based measure capturing broad-based
uncertainty in real time series. While statistically demanding to construct, it provides
useful short-term signals for tail-risk assessment in macro-financial conditions.
4 Correlation analysis identifies potential redundancy between series. Granger causality tests detect
predictive links with GDP and financial stress. PCA reduces dimensionality by extracting uncorrelated
components. Hierarchical clustering groups indicators with similar dynamics, and LASSO-based
variable selection identifies those groups with the strongest predictive power while penalising
overfitting.
5 The indicators selected are shown between asterisks in Chart 1 in the main text of this report.
6 The sources of the indicators can be found in Table 1 in the main text of this report.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Additionally, the World Uncertainty Index (WUI) shows predictive strength at horizons
of up to one year.
Military conflict, war and infrastructure risk indicators – The Geopolitical Risk
(GPR) Index quantifies geopolitical tensions, representing uncertainty signals with
broad application in empirical studies. However, it does not show high predictive
power for economic activity, possibly because its data origins lie in the Anglo-Saxon
news sources (see Box A for a discussion). In contrast, the Local Perception of
Geopolitical Risk (BiGPR) Indicator is a news-based index using local-language
media to measure how geopolitical events and tensions are perceived domestically.
Its predictive strength is most evident at horizons of one year and longer in the pre-
COVID sample, with some importance for shorter horizons. The National Security
Policy Equity Market Volatility (EMVsec) Tracker is a subcomponent of the Equity
Market Volatility Tracker and is based on news articles about national security
issues, identified through correlations with market volatility measures. It shows more
relevance at one-year horizons than in the short term. The Significant Cyber
Incidents (NoCyb) Indicator is representative of cyber-related geopolitical risks.
Although its empirical use is limited owing to its short time series, it captures an
increasingly important dimension of modern conflicts.
Trade-related uncertainty and disruption indicators – The Trade Policy
Uncertainty (TPU) Index measures the share of newspaper articles linking trade
policy with terms reflecting uncertainty and is a strong predictor at the one-year
horizon. The Global Supply Chain Pressure Index (GSCPI), a composite measure of
the transport costs and supply chain components of the Purchasing Managers’ Index
(PMI) showed predictive relevance only when the COVID-19 period was excluded
from the estimation sample. The Trade Openness (TO) Indicator is a widely used
official statistics measure that reflects both the vulnerabilities of countries to
geopolitical events and the reactions to these events.
Capital and financial market uncertainty indicators – The JLN Financial
Uncertainty (FinUn) Index is a model-based measure of forward-looking uncertainty
in financial series, with predictive value at horizons of up to one year. The Common
Volatility (COVOL) Index, derived from a factor model of volatility across asset
classes, has weaker but still notable relevance at quarterly horizons, reflecting its
sensitivity to abrupt global risk repricing. The first-quarter differences in the Financial
Fragmentation (FinFrag) Index showed some predictive strength at the one-year
horizon.
Politics & societal indicators – The Migration Fear (MigFear) Index is a news-
based measure of public concern over immigration, with predictive relevance up to
eight quarters ahead. The news-based Sanctions Intensity Index (GSI) provides
signals over the same horizon. Additionally, the Elections and Political Governance
EMV Tracker (EMVelec) shows predictive strength up to one year ahead.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Chart A
Main indicators used in the EU geopolitical risks analysis framework by category
a) General b) Military & infrastructure
(z-score) (z-score)
c) Trade d) Capital & finance
(z-score) (z-score)
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
e) Politics & societal f) Domestic (regional)
(z-score) (z-score)
Source: The sources corresponding to each indicator are given in Table 1 of the main text of this report.
Notes: EMV stands for equity market volatility. The charts represent the z-scores of the indicators concerned.
Current state of geopolitical risks
The indicators selected serve to provide an initial assessment of the current
state of the geopolitical risks to which EU Member States are exposed. The
assessment is conducted using the set of indicators selected and visual tools to
compare the latest developments with earlier geopolitical episodes or with specific
historical events.
The combination of indicators in a heatmap reveals a gradual intensification of
geoeconomic fragmentation since the global financial crisis and a surge in
policy uncertainty measures during 2024 and 2025. (Figure 4). Organised by risk
category, the heatmap facilitates comparative monitoring across the selected
indicators, helping identify elevated risks and thematic clusters across time. It allows
users to detect concentration of risks, shocks, emerging trends and persistent
structural vulnerabilities, depending on the profile of each indicator. Only indicators
with sufficient data coverage and empirical relevance – as identified in Sections
and Box B – are retained in the heatmap. The heatmap reveals a significant
intensification of geopolitical risks over the last decade, whereby measures of policy
uncertainty have especially surged during 2024 and 2025, driven primarily by a
sharp increase in global economic and trade policy uncertainty.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Figure 4
Heatmap of geopolitical indicators for EU geopolitical risks analysis by framework
category
Source: The sources corresponding to each indicator are given in Table 1.
Notes: EMV stands for equity market volatility. The indicators are normalised to a range [0;1]. The indicators displayed in the heatmap
are chosen for their statistical properties, timeliness, and global scope. The colour shading reflects their intensity, based on normalised
values. Instead of the Financial Fragmentation (FinFrag) common factor, the heatmap reports the average of its main (standardised)
components, the Foreign Direct Investment (FDI) Ratio and the Financial Flow Ratio. The Bilateral Indicator of Local Perception of
Geopolitical Risk (BiGPR) is not included, given that it captures country-specific perceptions rather than regional or global risks but is
retained in the recommended indicators owing to its empirical relevance. The indicators are winsorised (1-99 percentiles) to avoid
distortions induced by extreme outliers. The latest observations are for the second quarter of 2025 (dashed line on the right-hand
side).
While the geopolitical indicators heatmap captures only a limited subset of
long-term trend indicators, a broader set is shown in Chart 1. Political
fragmentation, for example, was already evident in the early 2000s following the
terrorist attacks in the United States and the subsequent global war on terror. The
global financial crisis of 2008 marked a turning point, with a reversal of globalisation
processes and a sustained rise in a range of indicators, reflecting financial and
mobility fragmentation, sanctions intensity, armed conflict and global economic policy
uncertainty. Some of the geopolitical fragmentation pressures accelerated further
with the escalation of trade tensions between the United States and China in 2017.
By 2025, many trend indicators stand at or near their historical peaks of the last
three to five decades, although the domestic SRI contrasts with this picture by
pointing to a decline in financial imbalances.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Chart 1
Changes in the geopolitical trend indicators selected for the EU geopolitical risks
analysis framework
Source: The sources corresponding to each indicator are given in Table 1.
Notes: Indicator values are normalised to a range [0; 1] based on the values from 2000 to the latest available data. The chart shows
annual or fourth quarter values for the year concerned.
In contrast to the fragmentation trend indicators, volatility and uncertainty
indicators make it possible to clearly identify the various events that have
shaped the pace and direction of geopolitical trends. Abrupt and widespread
economic shocks, such as the 2008 global financial crisis or the pandemic in 2020,
are clearly identified by the more general indicators, such as the financial and
macroeconomic uncertainty indices or the Common Volatility (COVOL) Index.
Outbreaks of major armed conflicts, by contrast, are better identified by spikes in the
GPR indicators or in the National Security Policy EMVsec Tracker, corresponding to
the Gulf War in 1991, the 9/11 terrorist attacks, the Iraq War in 2003 and the Russian
invasion of Ukraine in 2022. this invasion is still ongoing, as are Israel’s conflicts with
Hamas and Iran, the GPR Index has remained high (close to the 90th percentile of
the historical distribution).
More recently, indicators of global uncertainty, especially those capturing
trade uncertainty, have spiked, driven by the significant shifts in US foreign
and domestic policies and their potential implications for the international
economic and financial system. Chart 2 illustrates how different indicators have
reacted to major geopolitical developments in recent years, highlighting those that
registered pronounced spikes during key events. From the second quarter of 2025
(the end of the sample period), most of these indicators were at or near multi-decade
highs. At the same time, the Financial Uncertainty (FinUn) Indicator stood close to
the 70th percentile of the historical distribution, suggesting that the heightened
uncertainty was also weighing on the predictability of financial market developments.
By contrast, the Real Economic Uncertainty (RealUn) Index and Asset Price Volatility
Index, appear to be unaffected as yet by recent geopolitical events, both indices
remaining below their historical medians.
Financial stability risks from geoeconomic fragmentation
Assessing geopolitical risks to EU countries
Chart 2
Response to significant events of the geopolitical risk indicators for the EU
geopolitical risks analysis framework
Source: The sources corresponding to each indicator are given in Table 1.
Notes: The indicator values are normalised to a range [0; 1] based on the values from 2000 to the latest available data. The chart
shows the maximum value of a given indicator in the particular event window.
Overall, the indicators point to a two-stage dynamic in globalisation and
fragmentation, represented by elevated geopolitical and policy uncertainty but
more contained volatility. The post-Cold War decades were marked by strong
global integration, with rapid growth in trade openness and cross-border investment.
Since the 2008 global financial crisis, however, this process has slowed, and in
some areas reversed, giving way to renewed fragmentation reinforced by recent
geopolitical and policy shocks. At the same time, uncertainty indicators have risen
sharply across several domains, reaching historically high levels in recent years,
while volatility measures have remained relatively subdued or have reverted quickly
after short-lived spikes. This configuration has coincided with persistently high
geopolitical risks, increasing the likelihood of tail-risk events materialising in the near
term (Section ). From a financial stability perspective, one mitigating factor is the
fact that cyclical systemic risks in the EU remain relatively subdued, as reflected in
the low readings of the financial cycle indicators in our dataset.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
4 Transmission of geopolitical risk
Geopolitical events vary significantly in their nature, magnitude and
transmission channels, which means their effects on the economy and on
financial stability are far from uniform. Geopolitical risk propagates through
familiar macro-financial channels – uncertainty, risk appetite and trade – but also has
certain distinctive features, such as increased downside risks (especially tail risk),
stronger fragmentation effects (sanctions, reshoring and supply chain
reconfiguration) and heterogeneity across countries and sectors. Moreover, this
specific type of risk can also act as an amplifier of pre-existing vulnerabilities,
although the recent multitude of geopolitical events has not led to such amplification.
This section analyses the impact of geopolitical shocks, measured by a range of
selected indicators (Section 3), on the real economy and on the financial system.
The transmission on inflation by Anttonen and Lehmus (2025) is covered in Annex
and the remaining analysis uses macro-econometric models focused on the time
dimension of the transmission mechanism. The findings combine results for the euro
area and for individual EU Member States to document the heterogeneity not only in
the different types of geopolitical shocks but also in cross-country responses.
Macro-financial transmission of geopolitical risk
Beyond the effects on inflation, the global macro-financial repercussions of
geopolitical shocks can be illustrated using a factor-augmented vector
autoregressive (FAVAR) model. A multi-country model, based on Metiu (2025),
makes it possible to study geopolitical shock transmission by providing a data-rich
environment. This enables detailed analysis of international transmission channels
while accounting for cross-border linkages among major advanced economies. The
framework can capture the interactions between the (GPR) Index and a number of
unobserved factors estimated from a comprehensive panel data set for G7 and euro
area
Adverse geopolitical shocks propagate to the global economy through a
financial transmission channel, particularly by raising uncertainty and risk
aversion in global financial markets. An adverse geopolitical shock – an
unexpected increase in the GPR Index – results in a significant decrease in stock
prices, an increase in equity market volatility and a widening of corporate credit
spreads (Chart 3). Equity volatility returns to the pre-shock level relatively quickly,
whereas the impact on stock prices is more persistent.
7 Details to the model specification, the data and variable definitions can be found in Annex 2.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
Chart 3
International effects of an adverse geopolitical shock
Impulse response functions after a shock increase in the Geopolitical Risk Index across the
G7 and for selected EU Member States
(measure)
a) Stock prices b) Equity volatility c) Corporate credit spreads
d) EPU e) Business confidence f) Exports
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
g) Oil and gas production h) Sales i) Industrial production
j) Credit to PNFS k) Global shipping costs l) Consumer prices
Sources: Banque de France, Bloomberg Finance ., Federal Reserve Bank of St. Louis, United States Energy Information
Administration, Caldara and Iacoviello (2022), Gilchrist and Zakrajsek (2012), the Economic Policy Uncertainty Index by Baker, Bloom
and Davis (2016) and ESRB calculations.
Notes: EPU stands for Economic Policy Uncertainty Index and PNFS for private non-financial sector, The blue lines denote the
impulse responses (GDP-weighted averages across the G7 and selected euro area countries including Austria, Belgium, Finland,
Ireland, Netherlands, Portugal and Spain) with blue shaded 90% confidence intervals. The shock is scaled to generate an increase of
index points in the Standard Geopolitical Risk Index (GPR) on impact, consistent with the rise observed during the Russian
invasion of Ukraine from January to February 2022. The red lines with circles show counterfactual impulse responses for a scenario in
which the global financial factor does not react to the geopolitical risk shock. The sample that spans the period from January 1990 to
November 2023.
Macroeconomic channels also play an active role in the propagation of
geopolitical shocks to the global economy. Specifically, an unexpected increase
in the GPR Index is followed by a significant increase in news-based measures of
economic policy uncertainty and a decrease in survey-based measures of business
confidence, weighing on the global economy from the demand side. Additionally, any
such shock dampens exports and reduces global oil and natural gas production,
reflecting weaker global demand. Taken together, these developments weigh on
private consumption, as captured by retail sales, industrial output and credit to the
private non-financial sector (PNFS). On the supply side, a sudden rise in geopolitical
risk results in higher international shipping costs, supply chain reconfiguration and
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
sanctions on commodities; these can feed into higher consumer prices, increasing
price volatility and creating inflationary
Disabling financial market responses weakens the macro-financial impact,
underscoring the key role of financial markets in propagating geopolitical risk
shocks. A counterfactual experiment assesses the impact of a geopolitical risk
shock while holding constant the endogenous movements in an unobserved global
factor derived from financial volatilities and Equity prices, volatilities, credit
spreads, business confidence, exports, and production respond significantly less to
geopolitical risk shocks when the global financial factor’s endogenous reaction is
suppressed, highlighting the amplifying effect of adverse financial market responses.
Scenario analyses indicates that the euro area may face worsening macro-
financial conditions from a variety of adverse geopolitical shocks. The following
scenario builds on a large-scale Bayesian vector autoregression (BVAR), similar in
nature to the estimated global FAVAR, to analyse the impact and transmission of
geopolitical shocks as observed during the Russian invasion of
Following an intensification of geopolitical tensions, such as the escalation of
an armed conflict, the scenario analysis reveals adverse effects across
financial and economic indicators (Chart 4). Specifically, a generic conflict
scenario characterized by heightened geopolitical risk is considered, drawing on the
historical behaviour of the GPR and COVOL indices during the Russian invasion of
In such a scenario, equity and oil prices decline, while credit to non-
financial corporates contract significantly. Short-term government bond yields
decrease in response to the initial economic downturn. Notably, real GDP in the euro
area is initially less affected than that of the United States, highlighting regional
differences in the impact of such a scenario. However, the initial negative effect in
real GDP in the United States is followed by a moderate upturn, while the euro area
faces a decline towards the end of the forecast horizon. Risk indicators, including the
VIX Index, the high-yield option-adjusted corporate bond spreads for the euro area
and the CISS for the euro area, all rose markedly, signalling elevated financial stress
(Chart 4).
8 Given the broad range of geopolitical events covered by the geopolitical risk index, demand-side and
supply-side effects overlap in the model results. On balance, the model results suggest that inflationary
effects more than offset deflationary effects.
9 The counterfactual experiment is implemented using the method proposed in Camba-Mendez (2012).
10 The following scenario focuses on a combined increase in the geopolitical risk indicator (GPR) and
volatility (COVOL) as observed during the Russian invasion of Ukraine. A separate scenario focussing
on an increase in Economic Policy Uncertainty as observed during COVID-19 can be found in Annex 2.
11 The scenario is constructed by estimating an AR(1) model with an intercept for the GPR Index and
COVOL Index and the residual from the one-quarter-ahead forecast for the first quarter of 2022 – using
data up to the fourth quarter of 2021 – calibrates the generic conflict scenario. This shock is added to
the forecast for t+1 (first quarter of 2025), with subsequent quarters projected iteratively.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
Chart 4
Macro-financial implications after an increase in the Standard Geopolitical Risk Index
and COVOL Index
a) Euro Stoxx b) Brent spot oil price in EA c) Credit to non-financial
corporates in EA
(percentage changes) (percentage changes) (percentage changes)
d) Two-year benchmark
government yields in EA
e) Ten-year benchmark
government yields
f) Financial stress in EA
(difference) (difference) (difference)
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
g) Risky spread in EA h) Real Gross Domestic
Product in EA
i) HICP, all items in EA
(percentage, difference) (percentage changes) (percentage changes)
j) CBOE Volatility Index (VIX)
for S&P 500
k) Real Gross Domestic
Product in US
l) CPI-U: all items in US
(percentage changes) (percentage changes)
Sources: Caldara and Iacoviello (2022), Federal Reserve Bank of New York, Haver Analytics, Federal Reserve Economic Data and
ESRB calculations.
Notes: The blue lines denote the median scenario paths. The grey shaded areas show the 68% coverage interval. The red dashed
lines show the counterfactual median paths for a scenario in which the VIX Index does not respond throughout the scenario.
Shifting from a global to a regional focus, it is essential to evaluate how
geopolitical shocks affect the macro-financial environment in EU countries.
Bearing in mind the diverse nature of geopolitical risks (terrorist attacks, wars,
tensions, etc.), a high degree of heterogeneity in shock transmission is clearly to be
expected, closely related to each country’s geographic, economic, financial and
strategic position. It is possible to illustrate this impact within a multi-country vector
autoregression (VAR) framework. First of all, VAR models are estimated country-by-
country for 15 EU Member States using an unbalanced quarterly dataset and,
subsequently, the mean group estimator is applied by computing the GDP-weighted
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
cross-country Each country-specific VAR includes the GPR Index
(Caldara and Iacoviello, 2022) and relevant macroeconomic and financial
Geopolitical shocks are identified, using the same block-recursive
structure as in the FAVAR model, by assuming that they are contemporaneously
exogenous to macro-financial developments.
Chart 5 presents the average effect across EU Member States of an adverse
geopolitical shock, revealing a combined supply and financial
Elevated geopolitical tensions tighten aggregate supply conditions, reflected in
declining consumer confidence and rising inflation. The shock also increases CLIFS
stress indices, depresses house prices and lending, and raises borrowing costs for
non-financial enterprises. Overall, these responses signal a significant tightening of
financial conditions, which curtails corporate investment and activity, and heightening
macro-financial instability.
Chart 5
Effects of an adverse geopolitical shock on EU countries
Impulse response functions after an unexpected increase in the Standard Geopolitical Risk
Index across selected EU Member States
(standard deviations; quarters)
a) GPR b) Consumer confidence
12 Data are included for countries with at least 60 quarters of observations by the fourth quarter of 2024.
The countries included are Belgium, Germany, Greece, Spain, France, Italy, Latvia, Luxembourg,
Malta, the Netherlands, Austria, Portugal, Slovenia, Slovakia and Finland. Data from the first quarter of
2020 to the fourth quarter of 2020 are excluded to avoid distortions from the COVID-19 pandemic
(Lenza and Primiceri, 2022). In an unbalanced panel, countries with shorter samples have less precise
estimates.
13 For each country, the model specification comprises the following variables: an index of consumer
confidence as a forward-looking measure of economic fluctuations; inflation measured by the year-on-
year change in the GDP deflator; the Country-level Index of Financial Stress (CLIFS); the log of equity
prices deflated by consumer prices; the log of residential real estate prices deflated by consumer
prices; the log of the volume of credit to non-financial corporations; a composite measure of non-
financial corporates’ borrowing costs (a volume-weighted average of the interest rates on new short-
term and long-term loans capturing repricing and composition effects in new lending). The VAR is
estimated by ordinary least squares (OLS) with two lags for each country. To facilitate comparison
across countries, the data are standardised before VAR estimation.
14 The analysis omits the distance to the source of geopolitical risk, which may be a potential limitation.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
c) Inflation d) CLIFS
e) Equity prices f) RRE prices
g) NFC credit vol. h) NFC borrowing cost
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Notes: GPR stands for geopolitical risk, Std dev for standard deviation, CLIFS for the Country-Level Financial Stress Index, RRE for
residential real estate and NFC for non-financial corporation. These charts show the mean group impulse responses of macro-financial
variables to a one-standard deviation positive geopolitical risk shock (black solid lines). The mean group impulse responses are
calculated as the GDP-weighted average of the impulse responses for individual countries. The red shaded areas denote the GDP-
weighted average of 68% bootstrap confidence intervals to account for cross-country heterogeneity. Equity prices, RRE prices, and
NFC credit volumes are in real terms.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
The Russian invasion of Ukraine is one example of such a shock,
demonstrating how geopolitical escalation can rapidly cause real economic
losses and amplify financial system vulnerabilities. For instance, Russian gas
supply cuts and fears of a complete stoppage drove up gas prices, further fuelling
inflation in the euro area in 2022 and 2023. In response, monetary policy
implemented steep interest rate hikes, resulting in tighter financing conditions.
There is considerable cross-country heterogeneity in the estimated impact of
geopolitical shocks (Table 3). The magnitude and persistence of these effects
differed across countries. However, a model-consistent grouping relies on the
confidence (demand) and borrowing costs (supply) responses. Countries with large
declines in confidence and sharp increases in borrowing costs exhibited stronger
credit contraction (Belgium, Italy, the Netherlands, Greece and Austria), which may
be related to the risk repricing or expectations channel that tightens financial
conditions. By contrast, countries with muted confidence movements and limited cost
pass-through showed only modest (GDP-weighted average) credit declines
(Germany, France, Portugal, Slovenia and Slovakia). Nevertheless, the
heterogeneity persisted. A few small open economies – generally those with deeper
and more integrated financial markets, such as Belgium, the Netherlands and Austria
– were particularly vulnerable to such shocks, whereas larger economies, such as
Germany and France, as well as several smaller countries, including Portugal,
Slovakia, and Slovenia, appeared to be less affected. Another large country, Spain,
and a number of smaller countries, such as Malta, Latvia and Finland, fell
somewhere in between.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
Table 3
EU country-level effects of an adverse geopolitical shock
Minimum/maximum impulse responses over a 12-quarter period
Confidence Inflation CLIFS Equity prices RRE prices NFC credit
vol.
NFC borrowing
costs
BE -- ++ ++ - - - ++
AT -- ++ ++ - - 0 ++
NL -- ++ ++ - - 0 ++
LU - ++ ++ 0 - 0 ++
GR -- + + 0 - - +
IT - ++ ++ + - 0 ++
LV -- + + - - 0 +
ES - ++ + 0 - 0 +
FI -- + + 0 - 0 +
MT - + + - - 0 +
FR - + + 0 - 0 +
PT - + + 0 - 0 +
SI - + + 0 - 0 +
DE - 0 + - 0 0 +
SK 0 + + 0 0 0 +
Note: CLIFS stands for Country-Level Financial Stress Index and RRE for residential real estate. This figure shows the country-level
minimum/maximum impulse responses over a 12-quarter period of key macro-financial variables to a one-standard-deviation positive
geopolitical risk shock. "+/++" indicates a higher or much higher response than in other EU Member States. "−/−−" indicates a lower or
much lower response. "0" indicates that the impact was generally similar to that of other EU Member States. The red and orange
shading indicates the economically expected negative effect of an adverse geopolitical shock, while green indicates a positive (outlier)
effect.
The evidence presented here shows that geopolitical shocks can profoundly
impact the global macro-financial environment. Such shocks can propagate
through both financial and macroeconomic channels, weighing on the real economy
by weakening both demand and supply. Moreover, geopolitical shocks worsen
financial conditions globally as well as in the EU, by raising financial stress levels,
increasing borrowing costs and limiting credit to firms. This underscores the
importance of incorporating geopolitical risk into financial stability assessments,
focusing in particular on cross-country heterogeneity.
Tail risks to macro-financial conditions in the EU
Geopolitical risks may also affect the tails of macro-financial variables. This
section focuses on the effects of geopolitical shocks on the predictive distribution of
financial stress, systemic vulnerabilities and economic activity across the euro area
and in individual EU Member States. To this end, two distinct quantile regression
approaches were used, namely the panel quantile regression (PQR) and the quantile
vector autoregression (QVAR), each with its own merits. The PQR accounts for non-
linear effects across the different estimated quantiles, while making it possible to
control for other potential drivers of tail risks. By contrast, the QVAR captures
possible direct and indirect interactions between model variables giving rise to
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
additional non-linear Across both regression approaches, the core set of
variables covered financial stress (the CISS developed in Holló et al., 2012),
systemic vulnerability (the SRI, established in Lang et al., 2019) and economic
activity (real GDP), with corresponding indicators available at the country level.
Growth-at-risk in the EU
Integrating geopolitical risk into the GaR framework is a natural approach to
assessing the impact of tail risks to economic growth. For this section, the GaR
model specification used by the ESRB to evaluate the macroprudential policy stance
(ESRB, 2021) is augmented with geopolitical risk indicators to obtain quantitative
implications for systemic risks at EU and country level.
To assess the implications of a wide range of indicators, the first step consists
in identifying, in an EU panel setting, the most relevant indicators for
forecasting tail risks to GDP growth. The indicators are selected based on the set
described in Section 3 and a variable by applying machine learning techniques and
a light gradient boosting machine (LGBM) This approach not only allows
for an accurate estimation of tail-risk projections but also identifies potential
indicators with early-warning properties, signalling shifts in geopolitical and macro-
financial conditions before they materialise in observed outcomes, thereby
enhancing the timeliness and forward-looking capacity of the monitoring framework.
The results of the machine learning panel variable selection exercise largely
confirm the geopolitical indicator selection outlined in Section 3. The most
important contributions to four-quarters-ahead economic growth forecast come from
the WTUI, the fragmentation indicators, the EPU Index and, to a lesser extent, the
GPR Index and the MigFear Index (see Annex). Apart from the geopolitical set of
indicators, the most relevant auxiliary variables are related to credit to households
and non-financial companies, consumer confidence, the current account balance
and other systemic risk indicator components (Lang et al, 2019). From an early-
warning perspective, the lack of materialised shocks from geopolitical risks so far
limits the ability to examine the real-time usefulness of indicators such as EPI or
WTUI to directly predict downturns, although their effects on the risk distribution is
becoming increasingly important.
15 The estimates of the quantile regression rely on the seminal work by Koenker and Bassett (1978) and
have been applied by Adrian et al (2019) among others. The QVAR specifications rely on methodology
in Chavleishvili and Manganelli (2019), Bochmann et al. (2023) and Schüler (2025).
16 The LGBM is used owing to its advanced features, such as handling missing data, and custom loss
options, such as quantile loss. It is a fast, efficient, and scalable gradient boosting algorithm used for
regression tasks - it builds an ensemble of decision trees in a leaf-wise (best-first) manner and is
designed specifically for speed and accuracy for efficient and scalable machine learning tasks. The
best variables are selected based on a quantile loss and the empirical coverage measures, together
with the Shapley values to rank the best performing indicators by their marginal contribution to the
outcome of the model.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
The set of indicators used in the benchmark specification for the EU GaR models is
similar to those proposed in ESRB (2021),17 . augmented by the geopolitical risk
indicators:
where 𝑆𝑅𝐼𝑖,𝑡 is the Systemic Risk Indicator that tracks (cyclical) systemic risks in EU
Member States,18 𝐶𝐿𝐼𝐹𝑆𝑖,𝑡 is the Country-Level Index of Financial Stress and 𝐺𝐸𝑂𝑖,𝑡
represents the different geopolitical risk indicators For the estimation,
one geopolitical risk indicator per topical category is used to explore GDP growth tail
risks within the ESRB GaR framework, namely: (i) the EPU Index for general
uncertainty; (ii) the GPR Index for military/infrastructure; (iii) the WTUI Index for
trade; (iv) the COVOL Index for capital & finance; and (v) the MigFear Index for
politics & society.
Estimating the EU panel GaR model yields some heterogeneous results in terms of
coefficient plots –some of the indicators, such as the EPU Index and TPU Index
(Annex 2, Chart A2), have intuitive shapes and signs for the post-2014 period
(negative and upward-sloping). In accordance with a priori beliefs, the post-2014
results exhibit a clearer upward tendency and more pronounced negative effects in
the tails of the growth distribution, signalling potential threshold effects partially
overlapping with the onset of a financial fragmentation trend.
17 The empirical methodology aligns with the ESRB (2021) framework to provide policymakers with
quantitative metrics for assessing policy adequacy against risks and resilience. The approach
employed in this paper differs in two ways. First, while the Expert Group's stance indicator measures
the gap between the 50th and 10th percentiles of GDP growth, the current method studies the full
distributions and emphasizes individual percentiles, particularly the 10th percentile, to better capture
downside risks. Second, and also for this purpose, it focuses on an adjusted explanatory variable set:
for example, the policy dimensions is not included directly and the European Commission’s Economic
Sentiment Indicator (ESI) is added, which enhances the quantification of downside risks relative to
median predictions over shorter horizons (see Lang et al., 2022).
18 For robustness, both the Common Composite Indicator (CCI) and the Systemic Risk Indicator (SRI)
have been employed as measures of cyclical systemic risks, yielding similar results (see Annex 2).
19 The estimation includes country fixed effects to allow for country heterogeneity and make it possible to
estimate the results over a range of quantiles, while standardising the data to obtain more stable
results. This analysis emphasizes close-to-real-time monitoring with a four-quarter ahead forecasting
horizon. The societal and political indicator, deemed more relevant for medium-term forecasts (up to
eight quarters), has however been retained for the medium-term analysis, along with structural
indicators that play a significant role in cross-country comparisons.
𝑄𝑦𝑖,𝑡+4 = 𝑐𝑖 + 𝑦𝑖,𝑡 + 𝐸𝑆𝐼𝑖,𝑡 + 𝑆𝑅𝐼𝑖,𝑡 + 𝐶𝐿𝐼𝐹𝑆𝑖,𝑡 + 𝑆𝑅𝐼𝑖,𝑡 × 𝐶𝐿𝐼𝐹𝑆𝑖,𝑡 + 𝐺𝐸𝑂𝑖,𝑡 + 𝜀𝑖,𝑡 ,
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
Chart 6
EU growth-at-risk over time and across geopolitical indicators
Baseline EU growth-at-risk model versus models augmented with geoeconomic variables
a) GDP growth and GaR four quarters ahead b) GaR four quarters ahead for different
specifications and points in time
(percentages) (percentages)
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Note: GaR stands for growth-at-risk, ESI for Economic Sentiment Indicator, EPU for Economic Policy Uncertainty Index, GPR for
Global Geopolitical Risk Index, WTUI for World Trade Uncertainty Index and COVOL for Common Volatility Index. The augmented
models include geopolitical indicator variables. Panel a) displays the historical annual GDP growth rate alongside the median
prediction for four quarters ahead, based on the baseline specification. It also shows the range of 10th percentile predictions across
different specifications, including the baseline specification and specifications with geopolitical risk indicators, with each indicator (i to
v) iteratively added by category, as shown above. Panel b) shows the average growth-at-risk (. GDP growth at the tenth percentile)
expressed as annual percentage changes across various historical periods, estimated under the baseline specification and the six
specifications incorporating geopolitical risk indicators.
Tail risks to GDP growth from baseline and augmented models aligned on
average, but diverge during crises. The baseline model’s four-quarter forecast of
median GDP growth closely tracks actual outcomes (Chart 6, panel a), while the
10th percentile range of estimations (including geoeconomic variables) widen
significantly, particularly after 2015. This divergence highlights the growing
importance of geopolitical risks over the last decade and reflects the diverse nature
and contributions of those risks through time. Although the specifications converge
on average, notable deviations emerge during specific events, such as during the
2018-19 trade war (Chart 6, panel b), the disruptions induced by the COVID-19
pandemic and the recent geopolitical tensions. Incorporating geopolitical risk factors
into the estimations results in a more pronounced left tail of the GDP distribution,
appropriately capturing the geopolitical risks; the baseline specification without
geopolitical indicators could underestimate these risks in such environments.
Focusing on a single specification, the analysis resorts to the most promising
indicator from the general risk category, namely the EPU Index. This choice was
motivated by the convergence of multiple quantitative results presented throughout
this report: (i) the GaR LASSO selection procedure shows that the EPU Index is
especially relevant for predicting short- to medium-term risks (one-year ahead) for
the nine countries included in the sample (the detailed results are given in The
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
Annex); (ii) the machine learning approach ranked the EPU Index third among
geopolitical indicators (Annex 2, Chart A1) in terms of predictive power for the tails
of economic growth distribution; and (iii) the panel GaR specification exhibits a
significant increase in model fit when adding the EPU Index (Annex 2, Table 2).
Chart 7
Tail risks to EU GDP growth under EPU-augmented model
a) EU GaR and contributions of components b) Unconditional Distribution and Probability
density functions for EU growth
(percentage changes) (x-axis: percentage change, y-axis: probability density)
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Notes: GaR stands for growth-at-risk, SRI for Systemic Risk Indicator, CLIFS for Country-Level Index of Financial Stress, ESI for
Economic Sentiment Indicator and EPU for Economic Policy Uncertainty Index. Panel a) shows the quarterly GaR (expressed as an
annual percentage change), proxied by the change in the tenth percentile as estimated four quarters ahead under the EPU-
augmented specification, and broken down into the contributions of the explanatory variables included. Panel b) displays the
distribution of actual GDP growth alongside the predicted probability density functions (PDFs) for the two selected periods, derived by
fitting the quantile regression estimates from the geopolitical specification (the EPU).
The model specification incorporating the EPU Index predicts significantly
heightened tail risks to GDP growth post-2018. The breakdown highlights the fact
that the contribution made by economic policy uncertainty has become increasingly
negative since 2018, peaking during the COVID-19 period (2020) and persistently
dragging down the 10th percentile of conditional growth by % to % from 2022
onwards,20 emerging as the single largest contributor to downside risk (Chart 7,
panel a).
Slicing through the conditional growth distribution reveals the influence of
economic policy uncertainty on extreme negative outcomes in future GDP
growth. The shifts in the GDP growth distribution were particularly noticeable during
trade-related turbulence (Trump , 2018-19) and in the last two quarters of 2024
(Chart 7, panel b). While the risks appeared more balanced historically, they tilt to
the downside in the fourth quarter of 2024, exceeding the levels observed in 2018-
19. Relative to the unconditional distribution, episodes of heightened trade
uncertainty show greater deviations from the median, with thicker tails, indicating
higher variance in projected GDP growth and uncertainty. In late 2024, the
20 Additional examples are provided in the Annex 2.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
projections signalled an increased likelihood of GDP contraction, alongside a
downward shift in the median and upper range, compared with 2018-19.
Economic policy uncertainty shocks have a pronounced negative effect on
GaR, particularly when vulnerabilities are elevated. Chart 8 shows the estimated
impact on annual GDP growth of a one-standard-deviation shock to economic policy
uncertainty both across the key percentiles (Chart 8, panel a) and conditional on the
SRI being positive or negative (Chart 8, panel b). The impact of economic policy
uncertainty shocks is concentrated in the 10th percentile, emphasising that these
shocks primarily amplify downside risks. GDP growth shows a modest negative
response to economic policy uncertainty shocks on impact, which intensifies over
four quarters before gradually diminishing. This negative impact is stronger during
financial risk build-up phases (. when the SRI is positive), exposing greater
economic vulnerability. By contrast, low levels of systemic risk partially mitigated the
economy's sensitivity to such shocks. This suggests that (cyclical) systemic risk
dynamics may act as a structural driver and contribute to the build-up of
vulnerabilities to policy uncertainty shocks, with clear amplification effects observed
for adverse growth outcomes. In this regard, the key ingredients of a potential crisis
episode – underlying vulnerabilities, such as excessive credit or asset price growth,
and triggers such as heightened policy uncertainty – may interact negatively,
ultimately leading to significant real-economy effects, that is, the materialisation of a
tail event.
Chart 8
Quantile impulse responses of EU GDP growth quantiles to a one-standard-deviation
economic policy uncertainty shock across percentiles
a) EU GDP quantile impulse response b) EU GDP quantile impulse response
conditional on the SRI sign
(percentage points; impulse response time horizon in quarters) (percentage points; impulse response time horizon in quarters)
Source: ECB/ESRB workstream on financial stability risks from geoeconomic fragmentation.
Notes: IRF stands for impulse response function, EPU for Economic Policy Uncertainty Index and SRI for Systemic Risk Indicator. The
quantile IRFs are estimated using local projections, applying the methodology proposed in Adrian, T., F. Grinberg, N. Liang, S. Malik,
and J. Yu. 2022. "The Term Structure of Growth-at-Risk", American Economic Journal: Macroeconomics 14 (3): 283–323 (2022). The
quantile panel model used is the ESRB benchmark augmented with the EPU Index.
Financial stability risks from geoeconomic fragmentation
Transmission of geopolitical risk
EU geoeconomic fragmentation and financial cycles
The co-movement of macro-financial variables across countries influences
systemic risk in the euro area and affects the effectiveness of stabilisation
policies. The co-movement is crucial for deciding how stabilisation policies need to
be coordinated across countries to counter the impact of shocks. For example, when
financial cycles diverge across countries, national macroprudential tools may help
stabilise domestic conditions. A common monetary policy would, however, tend to
transmit unevenly, attenuating the imbalances in some countries while amplifying
those in others. By contrast, if financial cycles become too synchronised,
vulnerabilities may build up simultaneously across countries, generating amplifying
spillover ris
